In vivo examination of the local inflammatory response after implantation of Ti6Al4V samples with a combined low-temperature plasma treatment using pulsed magnetron sputtering of copper and plasma-polymerized ethylenediamine
Copper (Cu) could serve as antibacterial coating for Ti6Al4V implants. An additional cell-adhesive layer might compensate Cu cytotoxicity. This study aimed at in vitro and in vivo evaluation of low-temperature plasma treatment of Ti6Al4V plates with Ti/Cu magnetron sputtering (Ti6Al4V–Ti/Cu), plasma-polymerized ethylenediamine (Ti6Al4V–PPEDA), or both (Ti6Al4V–Ti/Cu–PPEDA). Ti6Al4V–Ti/Cu and Ti6Al4V–Ti/Cu–PPEDA had comparable in vitro Cu release and antibacterial effectiveness. Following intramuscular implantation of Ti6Al4V–Ti/Cu, Ti6Al4V–PPEDA, Ti6Al4V–Ti/Cu–PPEDA and Ti6Al4V controls for 7, 14 and 56 days with 8 rats/day, peri-implant tissue was immunohistochemically examined for different inflammatory cells. Ti6Al4V–PPEDA had more mast cells and NK cells than Ti6Al4V, and more tissue macrophages, T lymphocytes, mast cells and NK cells than Ti6Al4V–Ti/Cu–PPEDA. Ti6Al4V–Ti/Cu had more mast cells than Ti6Al4V and Ti6Al4V–Ti/Cu–PPEDA. Results indicate that PPEDA-mediated cell adhesion counteracted Cu cytotoxicity. Ti6Al4V–Ti/Cu–PPEDA differed from Ti6Al4V only for mast cells on day 56. Altogether, implants with both plasma treatments had antibacterial properties and did not increase inflammatory reactions.
Clinical application of implant materials requires a sufficient level of biocompatibility as well as specific physical and chemical properties suitable for the intended application. As established by experiences from its clinical use for bone replacement, titanium (Ti) and its alloys like Ti6Al4V have a very good biocompatibility and excellent mechanical strength. Ti-based implants were found to fulfil their intended function for ten or more years after implantation . There is however still room for their improvement. Among the most serious implant-related problems are bacterial infections which are the primary cause of implant failure [2, 3]. The most prevalent bacteria in this context are the gram-positive species Staphylococcus aureus and Staphylococcus epidermidis [4, 5]. Regarding this, the layer of plasma proteins formed on the implant surface after implantation as part of the implantation-related host reactions is of particular importance as it renders the implant surface susceptible for bacterial colonization and the formation of bacterial biofilms .
One possibility to improve the resistance against infections is the modification of the surface. A number of different coatings, among them for example noble metals like silver, antibiotics, other organic or inorganic antimicrobial agents, adhesion-resistant coatings, antibacterial bioactive polymers or nitrogen-monoxide delivering coatings, have been examined to achieve an antibacterial implant surface . As an alternative approach, we evaluated a low-temperature plasma based surface treatment called dual High Power Impulse Magnetron Sputtering (dual-HiPIMS) of copper (Cu) and found it to result in surfaces (Ti/Cu) with antibacterial properties causing a reduction of planktonic and biofilm-attached S. epidermidis and S. aureus bacteria . The underlying mechanism is the release of Cu ions from the Ti/Cu layer which was sputtered onto the implant surface. However, Cu ions are also toxic to mammalian cells in a concentration-dependant manner. Therefore, it is important to control the level of Cu release to reduce adverse effects against the peri-implant tissue which might affect implant ingrowth while on the other hand maintaining the antibacterial properties.
A possible solution could be the treatment of Ti/Cu-coated samples with an additional layer which could compensate for the Cu-related cytotoxicity. Ideally, this additional coating should possess bioactive properties itself for regulation of the tissue-surface interactions. In previous studies, we have examined a surface treatment with a process called Plasma Polymerized Ethylenediamine (PPEDA), resulting in a thin amino-group rich positively charged film on the Ti6Al4V surface (Ti6Al4V–PPEDA) which was found to increase adhesion and spreading of the human osteoblast cell line MG-63 in vitro . Thus, a PPEDA coating could be suitable as an additional layer on the surface of Ti/Cu-coated implants.
One important factor for the in vivo biocompatibility of a biomaterial which is also affected by bacterial infections is the inflammatory response after implantation. This has for example been demonstrated in an earlier study regarding the influence of complete Freund’s adjuvant, containing heat-killed mycobacteria and used to simulate a bacterial infection, on the immune response against polyester-based vascular prostheses . The acute and chronic inflammation following implantation is of major relevance for the short- and long-term stability and functionality of the implant. Most important among the cells responsible for these reactions are macrophages and other phagocytic cells . T lymphocytes are also involved in implantation-related host reactions , although their exact role has not been clarified so far [12, 13]. Additionally, mast cells were found to mediate the acute inflammatory response after implantation , and recent data demonstrated the infiltration of natural killer (NK) cells in the context of particle-mediated periprosthetic inflammation .
The aim of the present study was to examine the short- and long-term inflammatory reactions against Ti6Al4V plates coated with either Ti/Cu or PPEDA alone or a combination of Ti/Cu with an additional PPEDA layer, in comparison to control Ti6Al4V samples, after intramuscular implantation in rats. As in previous studies [16, 17, 18], one implant from each of the four implant series was simultaneously implanted into each animal for intraindividual comparison of the inflammatory reactions against different implants to reduce the level of variability between different experimental animals. To examine the local inflammatory response, total and tissue macrophages, total T lymphocytes, MHC-class-II-positive cells, mast cells and activated NK cells in the peri-implant tissue were immunohistochemically stained and quantified by digital image analysis.
2 Materials and methods
2.1 Implant samples and preparation
Ti6Al4V plates with a size of 5 × 5 mm and a thickness of 1 mm (DOT GmbH, Rostock, Germany) were used as implant samples.
2.1.2 Preparation of Ti/Cu thin films
Ti6Al4V plates with Ti/Cu thin films (henceforth designated Ti6Al4V–Ti/Cu implants) were prepared by dual-HiPIMS as previously described [7, 19]. Briefly, argon (Ar) ions, providing the sputtering effect, were accelerated towards Cu and Ti targets/cathodes at HiPIMS conditions with a repetition frequency of 100 Hz and pulse widths of 100 μs. All experiments were carried out in an Ar atmosphere at a pressure of 3 Pa (Ar flow rate 20 sccm). The dual-HiPIMS discharge was operated with typical mean discharge currents of 400 mA for Ti and 10 mA for Cu. Due to the pulse operation, the peak parameters reached much higher values with peak discharge current densities of about 2.5 A/cm2 for Ti and 0.1 A/cm2 for Cu. The high power densities caused intensive ionization of sputtered metal particles which also resulted in high ion flux. The different discharge currents for the sputtering of Ti and Cu were applied to eliminate the effect of different sputtering yields of Cu and Ti with a ratio of about four for Ar ion energies of 500 eV.
2.1.3 Preparation of PPEDA films
Plasma enhanced chemical vapour deposition (PECVD) for synthesis of thin films  was used for application of a mixture of Ar and the precursor ethylenediamine (Rotipuran®, ≥ 99.5 %, C2H8N2; Carl Roth, Karlsruhe, Germany) as processing gas to deposit PPEDA thin films (15–80 nm) on the Ti6Al4V plates by means of low pressure capacitively coupled radio frequency (RF) plasma at 13.56 MHz . The samples (henceforth designated Ti6Al4V–PPEDA implants) were placed on the powered RF electrode. The applied plasma processing parameters are the total pressure of 60 Pa (Ar:EDA = 5:1) at total gas flow rate of 24 sccm, the RF power of 60 W pulsed with the frequency of 10 Hz and a duty cycle of 50 % .
2.1.4 Preparation of combination of Ti/Cu and PPEDA thin films
In addition to the samples which were treated with either Ti/Cu or PPEDA alone, Ti6Al4V plates which received a Ti/Cu thin film subsequently followed by a PPEDA thin film, with both plasma processes performed as described above, were prepared (henceforth designated Ti6Al4V–Ti/Cu–PPEDA implants).
The implants from all four sample series were sterilized by gamma irradiation with a radiation dosage of 25 kGy.
2.2 Physico-chemical examination of plasma-treated samples
2.2.1 Examination of Ti/Cu thin films
The examination of the physico-chemical properties of the Ti/Cu thin film on the Ti6Al4V–Ti/Cu implants was described in detail elsewhere . Briefly, the chemical composition was examined by X-ray photoelectron spectroscopy (XPS) using a spectrometer (VG Microtech, W. Sussex, UK) equipped with a twin anode X-ray Mg Kα source and a hemispherical photoelectron energy CLAM-2 analyzer. Furthermore, the phase composition as well as the lattice parameters and domain sizes of the polycrystalline materials in the deposited films were evaluated with grazing incidence X-ray diffractometry (GIXD). X-ray reflectometry (XR) was used to examine density, thickness and roughness of the films. GIXD and XR investigations were both conducted with a Siemens D5000 diffractometer (Bruker AXS GmbH, Karlsruhe, Germany).
2.2.2 Examination of PPEDA thin films
The chemical composition of the PPEDA thin films on either the Ti6Al4V–PPEDA or the Ti6Al4V–Ti/Cu–PPEDA implants was examined by XPS as described for the Ti6Al4V–Ti/Cu implants. Furthermore, the PPEDA thin films were characterized by use of FTIR spectroscopy (Vertex 80v, Bruker Optic GmbH, Gremany) using IR reflection absorption technique for thin film analysis (Al substrate, parallel polarization, 75° incidence angle, spectral range 4,000–750 cm−1, resolution 0.25 cm−1). The thickness of the PPEDA thin films and the optical parameters were determined using a S2000 spectroscopic ellipsometer (Rudolph Research, Hackettstown, NJ, USA) within the spectral range between 300 and 750 nm at the incidence angle 70°.
2.3 Evaluation of Cu release from Ti6Al4V–Ti/Cu and Ti6Al4V–Ti/Cu–PPEDA implants
The release of Cu from Ti6Al4V–Ti/Cu and Ti6Al4V–Ti/Cu–PPEDA implants was measured by atomic absorption spectrometry (AAS) using a ZEEnit 650 spectrometer (Analytik Jena AG, Jena, Germany) with electrothermal atomisation. The samples were stored in 700 μl Dulbecco’s Modified Eagle Medium (DMEM; Invitrogen, Darmstadt, Germany), including 10 % Fetal calf serum (FCS Gold, PAA Laboratories GmbH, Pasching, Austria) and 1 % Gentamicin (Ratiopharm GmbH, Ulm, Germany), and incubated at 37 °C and 5 % CO2 atmosphere for different time periods. After the respective time points, the medium was exchanged with fresh DMEM, and the supernatant was acidified with nitric acid to stabilize the Cu ions and diluted to a concentration adequate for AAS analysis.
2.4 Examination of antibacterial effectiveness
2.4.1 Bacterial strain and inoculation conditions
The S. aureus strain ATCC 25923 was used in this study and grown either (without implant samples) in Caso-Bouillon (CB) (Carl Roth, Karlsruhe, Germany) or (with implant samples) in DMEM cell culture medium overnight at 37 °C under a 5 % CO2–20 % O2 atmosphere. The transition phase overnight CB culture was diluted 1:20 in CB and incubated to the late exponential growth phase for 6–8 h at 37 °C in a 5 % CO2–20 % O2 atmosphere. Bacterial cells were washed with PBS (pH 7.4), adjusted in DMEM to 1 × 108 colony-forming units (CFU)/ml and diluted to 1 × 102 CFU/ml. 1 ml of this suspension was used for inoculation of the experimental samples.
2.4.2 Examination of planktonic growth
Planktonic growth tests were performed in 24-well plates by inoculation with 1 × 102 CFU/well containing DMEM and Ti6Al4V, Ti6Al4V–Ti/Cu, or Ti6Al4V–Ti/Cu–PPEDA samples and incubation at 37 °C in a 5 % CO2–20 % O2 atmosphere for 1, 2, 4, 7 and 10 days. Subsequently, the culture supernatant was removed and the implant samples were carefully washed using 1 ml PBS. The washing buffer was pooled with the supernatant and centrifuged (4,000 rpm, 10 min, 4 °C, Heraeus Varifuge 3.OR; Kendro Laboratory Products, Osterode, Germany). The pooled samples were plated on CB agar plates and incubated for 24 h at 37 °C in an aerobic, CO2-enriched atmosphere for determination of the number of cultivable bacteria as CFU/ml.
2.4.3 Examination of biofilm formation
Bacterial biofilm formation on implant samples was examined using the PBS-washed samples from the experiments described above which were cautiously removed from the wells and transferred to a glass test tube containing 1 ml PBS. Low frequency ultrasonic treatment (Sonorex Digital 10P, Bandelin, Berlin, Germany) of the tubes for 5 min at 80 % intensity followed by vortexing for approximately 3 s was used to detach the biofilm from the sample surface. 100 μl aliquots of the resulting bacterial suspensions were plated on CB agar plates for determination of cultivable bacteria number as described above.
2.5 In vivo experiments
2.5.1 Laboratory animals
24 male Lewis rats (age 100 days, mean weight 311 ± 12 g) were kept in our in-house facilities under conventional housing and feeding conditions. All animal experiments were performed in full accordance with the animal protection law of the Federal Republic of Germany in its new version of 1 January 1987, with the principles of care for animals in laboratories (drawn up by the National Society for Medical Research) and with the Guidelines for Keeping and Using Laboratory Animals (NIH Publication No. 80-23, revised 1985).
2.5.2 Implantation procedure and tissue sampling
The animals were anesthetized by i.p. application of a mixture of Rompun® (Bayer, Leverkusen, Germany) and Ketamin® (Sanofi-Ceva, Düsseldorf, Germany). One sample from each of the four implant series was implanted in each animal into small intramuscular pockets in the neck musculature. The four implants were arranged in a rectangular formation and separated by at least 2 cm from each other, and fixed in their muscular pockets with a nonresorbable synthetic polypropylene suture (PROLENE®, Ethicon Endo-Surgery, Inc., Hamburg, Germany). Eight randomly selected animals were euthanized after 7, 14 and 56 days, respectively, and the implants with a sample of the surrounding tissue were carefully explanted after surgical opening of the implantation site. After freezing the samples immediately with laboratory freezer spray New Envi-Ro-Tech™ (Thermo Electron Corporation, Pittsburgh, USA), they were cut with a scalpel with the section plane at right angles with the implants followed by careful removal of the implants from the frozen tissue using tweezers. The embedding medium Shandon Cryomatrix™ (Thermo Electron Corporation, Pittsburgh, USA) was used to fill the remaining tissue pockets to preserve their form during further processing. Subsequently, the samples were shock frozen in liquid nitrogen and stored at −80 °C until further histological examination.
2.6 Morphological examination
2.6.1 Immunohistochemistry and histochemistry
For immunohistochemical examination, frozen sections with a thickness of 5 μm were prepared from the samples using a Cryotome 2800 Frigocut N (Reichert-Jung, Nussloch, Germany). The samples were stained with the following primary antibodies according to the respective manufacturer’s protocols: ED1 for monocytes and macrophages, ED2 for tissue macrophages, R73 for T lymphocytes, OX6 for MHC-class-II-positive cells (all obtained from MorphoSys AbD Serotec GmbH, Duesseldorf, Germany), AD1 for mast cells (BD Biosciences, Heidelberg, Germany) and ANK61 for activated NK cells (Santa Cruz Biotechnology, Heidelberg, Germany). Bound primary antibodies were detected using the Alkaline Phosphatase Anti-Alkaline Phosphatase method (APAAP; DakoCytomation GmbH, Hamburg, Germany). A nuclear counterstaining with Haematoxylin was performed according to established standard procedure.
2.6.2 Microscopic equipment
For microscopic evaluation of the stained histological samples, a light microscope CX41 (Olympus, Hamburg, Germany) together with a colour camera DP20 (1,600×1,200 Pixel, Olympus, Hamburg, Germany) at a magnification of 100× was used to obtain digital images.
2.6.3 Image analysis procedure
Using the image analysis program ImageJ v1.43 (U. S. National Institutes of Health, Bethesda, Maryland, USA)  and the software plugins Grid and CellCounter, positively stained cells in defined areas were counted in the microscopic images . Briefly, five representative squares with an area of 20,000 pixels per square in direct vicinity of the implants pockets were selected, giving a total analyzed area of 100,000 pixels per image which was reduced by the area of artefacts and other regions within a square which did not contain tissue if necessary. Using a microscopic slide with a printed length scale, it was determined that one pixel corresponded to an area of 0.4796 μm2 in the chosen microscopic magnification. The final results for all images are the averaged counts from two independent investigators and are provided as positively stained cells per μm2.
2.7 Statistical data analysis
The results for the Cu release experiments are presented as mean value ± SD (n = 3 for each analysis), and the unpaired t test was used to compare the results between the implants. The microbiological experiments were performed in triplicates. For morphometric analysis, the cell numbers in the peri-implant tissue of the four different implants on the same experimental day were compared using the non-parametric paired Wilcoxon signed rank test, and the results for each of the implant series over time were analyzed using the non-parametric unpaired Kruskal–Wallis test. A P value of less than 0.05 was considered significant for all tests. Statistical analysis was performed using the software system GraphPad Prism version 4.03 (GraphPad Software, Inc., San Diego, CA, USA).
3.1 Physico-chemical properties of plasma films
3.1.1 Ti/Cu thin films
The growth and the physico-chemical properties of the Ti/Cu thin films were described elsewhere . It was shown that even a very low Cu discharge current of 10 mA resulted in a strong contribution of Cu to the total chemical composition of the film which consisted of about 88–90 % Cu and roughly 10 % Ti. The film density as estimated by XR measurements was 6.85 g/cm3 and corresponded with chemical composition. Examination by GIXD revealed the formation of larger Cu grains with Ti incorporated into the films. The size of the crystal lattice parameter was estimated from the XR examination as about 0.44 nm.
3.1.2 PPEDA films
3.1.3 XPS surface analysis
X-ray photoelectron spectroscopy (XPS) examination of the elemental content of untreated Ti6Al4V surfaces (Ti6Al4V) and Ti6Al4V surfaces treated with dual High Power Impulse Magnetron Sputtering of copper (Ti6Al4V–Ti/Cu), Plasma polymerized ethylenediamine (Ti6Al4V–PPEDA) or a combination of both processes (Ti6Al4V–Ti/Cu–PPEDA)
16.3 ± 0.2
3.0 ± 0.1
2.5 ± 0.4
0.5 ± 0.1
44.4 ± 0.9
28.8 ± 0.9
4.1 ± 0.3
3.8 ± 0.2
32.4 ± 0.9
49.5 ± 1.3
66.2 ± 0.3
66.1 ± 0.1
29.5 ± 0.0
29.3 ± 0.3
18.7 ± 1.1
0.7 ± 0.1
3.2 Cu release from Ti6Al4V–Ti/Cu and Ti6Al4V–Ti/Cu–PPEDA implants
3.3 Antibacterial effectiveness
Numbers of planktonic and biofilm S. aureus bacteria as given in colony-forming units per ml (CFU/ml) growing on untreated Ti6Al4V surfaces (Ti6Al4V) and Ti6Al4V surfaces treated with dual High Power Impulse Magnetron Sputtering of copper (Ti6Al4V–Ti/Cu) or a combination of Ti/Cu with plasma polymerized ethylenediamine (Ti6Al4V–Ti/Cu–PPEDA) after inoculation with 1.0 × 102 CFU/ml on day 0
1.9 × 106
1.2 × 106
1.9 × 106
1.3 × 106
2.5 × 106
3.7 × 101
7.0 × 106
6.1 × 106
5.0 × 106
2.3 × 106
2.6 × 107
4.7 × 101
5.3 × 101
3.4 Morphological examination
3.4.1 Total monocytes/macrophages (ED1)
3.4.2 Tissue macrophages (ED2)
Comparable to the total monocytes and macrophages, a significant decline of the tissue macrophages over time was observed for all four implant series (Fig. 3b; Ti6Al4V: P = 0.0005; Ti6Al4V–Ti/Cu: P = 0.0075; Ti6Al4V–PPEDA: P = 0.0007; Ti6Al4V–Ti/Cu–PPEDA: P = 0.0017). The Ti6Al4V–Ti/Cu–PPEDA samples tended to have a lower count of ED2-positive cells than the Ti6Al4V samples and the Ti6Al4V–PPEDA samples on day 7, this differences were however not significant. A similar outcome was found on day 14 with significantly lower tissue macrophage numbers for the Ti6Al4V–Ti/Cu–PPEDA samples compared to the Ti6Al4V–PPEDA samples (P = 0.0078). No significant differences between the implants were found on day 56.
3.4.3 T lymphocytes (R73)
In general, the number of T lymphocytes was markedly lower than the number of total and tissue macrophages for all implant series throughout the study period. A significant decrease for the number of T lymphocytes from days 7 to 56 was found only for the Ti6Al4V–PPEDA samples and the Ti6Al4V samples (Fig. 3c; Ti6Al4V–PPEDA: P = 0.0041, Ti6Al4V: P = 0.0393). On day 14, the Ti6Al4V–Ti/Cu–PPEDA samples had a lower T lymphocyte number than the Ti6Al4V–PPEDA samples (P = 0.0078) while there were no significant differences between the four implant series on days 7 and 56.
3.4.4 MHC-class-II-positive cells (OX6)
The number of MHC-class-II-positive cells decreased significantly over time for all four implant series (Fig. 3d; Ti6Al4V: P = 0.0003; Ti6Al4V–Ti/Cu: P = 0.0002; Ti6Al4V–PPEDA: P = 0.0008; Ti6Al4V–Ti/Cu–PPEDA: P = 0.0004). No significant differences between the four implant series on any experimental were observed. Compared to day 56, the results for the MHC-class-II-positive cells in the early phase on days 7 and 14 were characterized by a higher individual variability.
3.4.5 Mast cells (AD1)
For the mast cells, a significant decrease from days 7 to 56 was found for the Ti6Al4V–PPEDA samples, the Ti6Al4V–Ti/Cu samples and the Ti6Al4V samples but not for the Ti6Al4V–Ti/Cu–PPEDA samples (Fig. 3e; Ti6Al4V: P = 0.0004; Ti6Al4V–Ti/Cu: P = 0.0002; Ti6Al4V–PPEDA: P = 0.0009). On day 7, the Ti6Al4V–Ti/Cu–PPEDA samples did not differ from the Ti6Al4V samples and had significantly lower numbers than the Ti6Al4V–Ti/Cu samples (P = 0.0078) and the Ti6Al4V–PPEDA samples (P = 0.0391), which both also had higher numbers than the Ti6Al4V samples (Ti6Al4V–Ti/Cu: P = 0.00156, Ti6Al4V–PPEDA: P = 0.00391). Similar results were found on day 14 on which the Ti6Al4V–Ti/Cu–PPEDA samples did not differ from the Ti6Al4V samples and had a significantly lower count of AD1-positive cells than the Ti6Al4V–PPEDA samples (P = 0.0156). In contrast, the number of mast cells on this day was significantly increased for the Ti6Al4V–Ti/Cu samples (P = 0.0078) and the Ti6Al4V–PPEDA samples (P = 0.00156) in comparison to the Ti6Al4V samples. On day 56, the Ti6Al4V–Ti/Cu samples (P = 0.0078) and the Ti6Al4V–PPEDA samples (P = 0.0234) were found to have significantly higher mast cell numbers than the Ti6Al4V samples.
3.4.6 Activated natural killer cells (ANK61)
For the activated NK cells, no significant change over time was found for any of the four implant series (Fig. 3f). While the Ti6Al4V–Ti/Cu–PPEDA samples and the Ti6Al4V–Ti/Cu samples did not differ significantly from the Ti6Al4V samples on any experimental day, the Ti6Al4V–PPEDA samples had a significantly higher number of activated NK cells than the Ti6Al4V samples (P = 0.0078) and the Ti6Al4V–Ti/Cu–PPEDA samples (P = 0.0156) on day 7. Furthermore, the Ti6Al4V–PPEDA samples also had a significantly higher count of ANK61-positive cells on day 14 in comparison to the Ti6Al4V–Ti/Cu–PPEDA samples (P = 0.0234) and to the Ti6Al4V–Ti/Cu samples (P = 0.0391).
Surface modification for biomaterials such as Ti and its alloys like Ti6Al4V offers the possibility to influence tissue-material interactions and thereby to improve the clinical performance of medical implants by reducing the incidence of implant-related complications, among which bacterial infections are prevalent and belong to the most serious. In this regard, it should also be noted that bacterial infections can aggravate immunological and inflammatory reactions. For example, this was demonstrated by the effects of application of Complete Freund’s Adjuvant, containing heat-killed mycobacteria, on the systemic immune response following implantation of vascular prostheses .
In previous in vitro experiments, a pronounced antibacterial effectiveness against S. epidermidis and S. aureus was found for Ti6Al4V implants modified with a low-temperature plasma based process called dual-HiPIMS of Cu . These Ti/Cu-coated samples (Ti6Al4V–Ti/Cu implants) were additionally equipped with an amino-group rich positively charged layer of PPEDA, resulting in samples with a combination of two layers (Ti6Al4V–Ti/Cu–PPEDA implants). The rationale behind this approach was to reduce possible adverse tissue reactions resulting from the known cytotoxicity of Cu, based on the observation that Ti6Al4V samples with a PPEDA film (Ti6Al4V–PPEDA implants) increased adhesion and spreading of human MG-63 osteoblasts in vitro . Therefore, the aim of the study was the in vivo evaluation of the short- and long-term local inflammatory response after implantation of Ti6Al4V–Ti/Cu, Ti6Al4V–PPEDA, Ti6Al4V–Ti/Cu–PPEDA and uncoated Ti6Al4V control samples in rats.
Following preparation of the different sample series, miscellaneous analytical methods were used to evaluate their physico-chemical properties. Most importantly of those results (Table 1), analysis of the elemental composition by XPS confirmed the successful coating of the Ti6Al4V surfaces with Ti/Cu or PPEDA films as well as the combination of both. Furthermore, in vitro evaluation revealed that the Ti6Al4V–Ti/Cu and the Ti6Al4V–Ti/Cu–PPEDA samples did not differ significantly regarding the Cu release at any time point up to 240 h. Accordingly, the microbiological evaluation demonstrated that there was no significant difference between these two sample series relating to their antibacterial effectiveness. On both, no viable planktonic or biofilm bacteria were observed after inoculation, in contrast to the Ti6Al4V control samples on which bacterial growth was observed until the end of the in vitro examination after 10 days.
Based on experience from previous in vivo studies [16, 17, 18], the model chosen in the present study was simultaneous intramuscular implantation of four different implants into the neck musculature of rats. The sample implantation and explantation are easy to perform in a uniform manner in this model, with a short OP duration as well as reduced surgical complications and post-operative stress being important factors from an animal welfare point of view. Furthermore, implantation into muscular tissue is well comparable to the clinical application of Ti6Al4V implants, and musculature is well supplied with blood which makes it especially appropriate for the evaluation of inflammatory reactions. The main advantage of simultaneous implantation is the direct comparison of inflammatory reactions against different implants within the same animal, thereby reducing the broad individual variability of host reactions as observed previously in studies on the immunological reactions against vascular prostheses [26, 27].
The explantation and immunohistochemical examination was performed after 7, 14 and 56 days, thereby encompassing short- as well as long-term reactions. Overall, the in vivo results indicate that the intensity of inflammation shifted from acute to chronic for all four implant series as evident from the declining reactions for the total and the tissue macrophages as well as the MHC-class-II-positive cells but also for the T lymphocytes and the mast cells. However, a number of differences between the implants were observed in the early as well as in the late study phase. Throughout the study, the Ti6Al4V–PPEDA implants caused a stronger mast cell response than the Ti6Al4V implants. Furthermore, on day 7 the number of activated NK cells was higher for the Ti6Al4V–PPEDA implants than for the Ti6Al4V implants. These results indicate that the increased cell adhesion as observed in vitro for the amino-group rich surface leads to short- and long-term adverse effects in vivo regarding attraction of inflammatory cells. As this was not found for Plasma Polymerized Allylamine (PPAAm) films which were evaluated in a previous study , it can possibly be attributed to the higher amino-group density of PPEDA compared to PPAAm due to the different chemistry of the respective precursor molecules (two amino groups for ethylenediamine, one for allylamine). A similar result regarding increased mast cell numbers over the whole study period was also observed for the Ti6Al4V–Ti/Cu implants, apparently caused by the known cytotoxicity of Cu.
Interestingly, the Ti6Al4V–Ti/Cu–PPEDA implants caused less pronounced inflammatory reactions in the early phase than either the Ti6Al4V–PPEDA implants or the Ti6Al4V–Ti/Cu implants. This was observed for total and tissue macrophages, mast cells, activated NK cells and T lymphocytes in comparison to the Ti6Al4V–PPEDA implants, and for mast cells and total macrophages in comparison to the Ti6Al4V–Ti/Cu implants. Furthermore, unlike the Ti6Al4V–PPEDA and the Ti6Al4V–Ti/Cu implants, the implants which received both plasma treatments were comparable to the Ti6Al4V implants, with a less pronounced response for tissue macrophages on days 7 and 14 as well as moderately higher numbers for the activated NK cells on day 7 and for the mast cells on day 56. A possible explanation regarding the differences between the Ti6Al4V–Ti/Cu–PPEDA implants in comparison to the Ti6Al4V–PPEDA and the Ti6Al4V–Ti/Cu implants could be that the contrary actions of Cu cytotoxicity on the one hand and the cell adhesive properties of the amino-group rich surface on the other hand counteract each other when both layers are combined.
As few studies have been published so far about the involvement of NK cells in the inflammatory response after implantation of biomaterials, the observation that they were increased for the Ti6Al4V–PPEDA samples compared to the Ti6Al4V–Ti/Cu–PPEDA samples on days 7 and 14 as well as compared to the Ti6Al4V samples on day 7 are particularly notable. In a study on subcutaneous implantation of sheep collagen in rats, NK cells were rarely detected , and no NK cells were found in another study following repetitive subcutaneous implantation of cross-linked collagens, polyurethane and silicone in rats . In contrast, a more recent study on fibrinogen adsorbed to chitosan films showed an increased adhesion of NK cells as well as an enhancement of their recruitment of mesenchymal stromal cells .
It is noticeable that the differences regarding the NK cell numbers in the present study are in line with the results for the tissue macrophages, the T lymphocytes and the mast cells, mostly observable for the comparison between the Ti6Al4V–PPEDA and the Ti6Al4V–Ti/Cu–PPEDA samples. This indicates that the NK cells could play a regulatory role through reciprocal interactions with macrophages and T lymphocytes as described previously for other situations . The discrepancies between the present results and the studies mentioned above could be due to different materials (Ti with bioactive thin films vs. collagens and synthetic polymers) and different implantation sites (i.m. vs. s.c.). Future studies could clarify the role of NK cells in biomaterials-related inflammatory reactions, for example whether they are predominantly CD56bright cells as this subset was shown to be enriched in inflammatory sites, amplifying and maintaining the inflammation .
Taken together with the in vitro data showing that the additional PPEDA film does not influence the Cu release from the Ti/Cu thin film or its antibacterial effectiveness, the in vivo results demonstrate that the combination of both plasma processes could be used to create implants which possess antibacterial properties and do not elicite an increased inflammatory response in comparison to Ti6Al4V controls. Future work to improve the outcome include further studies regarding the relationship between plasma process parameters and surfaces properties as well as development of a plasma process for creation of a single mixture layer within the same reactor instead of two separate layers.
We would like to thank Kirsten Tornow, Kathleen Arndt and Robert Mrotzek for their excellent technical support. This study was supported by the Federal Ministry of Education and Research (Grant No. 13N9779, Campus PlasmaMed).
- 1.Esposito M. Titanium for dental applications. In: Brunette DM, Tengvall P, Textor M, Thomsen P, editors. Titanium in medicine. Berlin: Springer; 2001. p. 172–230.Google Scholar
- 3.Arciola CR, Alvi FI, An YH, Campoccia D, Montanaro L. Implant infection and infection resistant materials: a mini review. Int J Artif Organs. 2005;28:1119–25.Google Scholar
- 4.von Eiff C, Arciola CR, Montanaro L, Becker K, Campoccia D. Emerging Staphylococcus species as new pathogens in implant infections. Int J Artif Organs. 2006;29:360–7.Google Scholar
- 6.Zhao L, Chu PK, Zhang Y, Wu Z. Antibacterial coatings on titanium implants. J Biomed Mater Res B Appl Biomater. 2009;91:470–80.Google Scholar
- 12.Rodriguez A, Voskerician G, Meyerson H, Macewan SR, Anderson JM. T cell subset distributions following primary and secondary implantation at subcutaneous biomaterial implant sites. J Biomed Mater Res A. 2008;85:556–65.Google Scholar
- 20.d′Agostino R. Plasma deposition, treatment and etching of polymers. Boston: Academic Press; 1990.Google Scholar
- 23.Gallagher SR. Digital Image Processing and Analysis with ImageJ. Curr Protoc Essen Lab Tech. 2010;3:A.3C.1–.24.Google Scholar
- 26.Wilhelm L, Zippel R, von Woedtke T, Kenk H, Hoene A, Patrzyk M, Schlosser M. Immune response against polyester implants is influenced by the coating substances. J Biomed Mater Res A. 2007;83:104–13.Google Scholar
- 27.Schlosser M, Zippel R, Hoene A, Urban G, Ueberrueck T, Marusch F, Koch A, Meyer L, Wilhelm L. Antibody response to collagen after functional implantation of different polyester vascular prostheses in pigs. J Biomed Mater Res A. 2005;72:317–25.Google Scholar
- 32.Dalbeth N, Gundle R, Davies RJ, Lee YC, McMichael AJ, Callan MF. CD56bright NK cells are enriched at inflammatory sites and can engage with monocytes in a reciprocal program of activation. J Immunol. 2004;173:6418–26.Google Scholar